期刊文献+

基于动作子空间和权重条件随机场的行为识别 预览

Behavior Recognition Based on Action Subspace and Weight Condition Random Field
在线阅读 免费下载
分享 导出
摘要 针对单目视频中的人类行为识别,提出了基于动作子空间与权重条件随机场的行为识别方法。该方法结合了基于特征提取的核主分量分析(KPCA)与基于运动建模的权重条件随机场(WCRF)模型。探讨了通过非线性降维行为空间的基本结构,并在运动轨迹投影过程中保留清晰的时间顺序,使人体轮廓数据表示更紧凑。WCRF通过多种交互途径对时间序列建模,从而提高了信息共享的联合精确度,具有超越生成模型的优势(如放宽观察之间独立性的假设,有效地将重叠的特征和远距离依存关系合并起来的能力)。实验结果表明,该行为识别方法不仅能够准确地识别随时间、区域内外人员变化的人类行为,而且对噪声和其他因素鲁棒性强。 For human behavior recognition in monocular video, a method for recognizing human behavior based on action subspace and weighted condition random field is presented in this paper. This method combines kernel principal component analysis (KPCA) based on feature extraction and weighted conditional random field (WCRF) based on activity modeling. Silhouette data of human is represented more compactly by nonlinear dimensionality reduction that explores the basic structure of action space and preserves explicit temporal orders in the course of projection trajectories of motions. Temporal sequences are modeled in WCRF by using multiple interacting ways, thus increasing joint accuracy by information sharing, and this model has superiority over generative ones (e.g., relaxing independence assumption between observations and the ability to effectively incorporate both overlapping features and long-range dependencies). The experimental results show that the proposed behavior recognition method can not only accurately recognize human activities with temporal, external and internal person variations, but also considerably robust to noise and other factors.
作者 王智文 蒋联源 王宇航 欧阳浩 张灿龙 黄镇谨 王鹏涛 WANG Zhi-wen1, JIANG Lian-yuan1,2, WANG Yu-hang3, OUYANG Hao1, ZHANG Can-long4, HUANG Zhen-jin1,2, WANG Peng-tao5(1. College of Computer Science and Communication Engineering, Guangxi University of Science and Technology Liuzhou Guangxi 545006;2. Guangxi Experiment Center of Information Science Guilin Guangxi 541004;3. Institute of Automobile and Traffic Engineering, Guilin University of Aerospace Technology Guilin Guangxi 541004;4. School of Computer Science & Information Technology, Guangxi Normal University Guilin Guangxi 541004;5. College of Electrical and Information Engineering, Guangxi University of Science and Technology Liuzhou Guangxi 545006)
出处 《电子科技大学学报》 EI CAS CSCD 北大核心 2017年第2期412-418,共7页 Journal of University of Electronic Science and Technology of China
基金 国家自然科学基金(61462008,61365009) 广西省自然科学基金(2013GXNSFAA019336,2014GXNSFAA118368,2016GXNSFBA380081)
关键词 人类行为识别 人体轮廓提取与表示 核主分量分析 非线性降维 权重条件随机场 human activities recognizing human silhouette extraction and representation kernel principalcomponent analysis (KPCA) nonlinear dimensionality reduction weighted conditional random field (WCRF)
作者简介 王智文(1969-),男,博士,教授,主要从事机器学习与计算机视觉、移动目标检测与识别方面的研究.
  • 相关文献

参考文献2

二级参考文献27

  • 1SHAWE-TAYLO J, CRISTIANINI N, Kernel methods for pattern analysis[M]. Cambridge: Cambridge University Press, 2004:15-25. 被引量:1
  • 2CAMPS-VALLS ~ SHERVASHIDZE N, BORGWARDT M K. Spatio-spectral remote sensing image classification with graph kernels[J]. IEEE Geoscience and Remote Sensing Letters, 2010, 7(4): 741-745. 被引量:1
  • 3LIU Xiao-fang. Research on remote sensing image classification methods based on kernel theory[D]. Chengdu: University of Electronic Science and Technology of China, 2011. 被引量:1
  • 4FRANC V, HLAV,/~t2 V. Greedy algorithm for a training set reduction in the kernel methods[M]. Heidelberg: Springer- Verlag Berlin, 2003. 被引量:1
  • 5FRANC V. Optimization algorithms for kernel methods[D]. Czech: Czech Technical University, 2005. 被引量:1
  • 6TANGKUAMPIEN T, SUTER D. Human motion de-noising via greedy kernel principal component analysis filtering [C]//Proc of the IEEE 8th International Conference on Pattern Recognition. Piscataway, USA: IEEE, 2006. 被引量:1
  • 7QIN Jian-ling, LI Jun. Theory of feature extraction and samples filtration based on KPCA[C]//Proe of the 12th International Conference on Industrial Engineering and Engieering Management. Beijing: Industrial Engineering Institute of China, 2005:1577-1580. 被引量:1
  • 8SCH(3LKOPF B, SMOLA A, MULLER K R. Nonline~x component analysis as a kernel eigenvalue problem[J]. Neural Computation, 1998(10): 1299-1319. 被引量:1
  • 9BANKJ M H, SHIRAZI A A B. New kernel function for hyperspectral image classification[C]// The 2rid International Conference on Computer and Automation En~,ineerin~,. Piscatawav: IEEE Comouter Society. 2010. 被引量:1
  • 10CRISTIANINI N, SHAWE-TAYLOR J. An introduction to support vector machines and other kernel-based learning methods[M]. Translated by LI Guo-zheng, WANG Meng, ZENG Hua-jun. Beijing: Publishing House of Electronic Industry, 2004. 被引量:1

共引文献2

投稿分析

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部 意见反馈